Estimating the travel distance of channelized rock avalanches using genetic programming method

A number of channelized rock avalanche events triggered by the Wenchuan earthquake resulted in considerable destruction. Identifying the factors affecting the travel distances of channelized rock avalanches and predicting their travel distances are crucial for mitigating disaster impacts. Considering the advantages of machine learning in data processing, 49 channelized rock avalanche events and three machine learning methods (genetic programming (GP), support vector machine (SVM), and multilayer perceptron (MLP)) were selected for model optimization and establishment. It was found that the GP model outperforms both SVM and MLP, achieving the best fit alongside superior predictive accuracy and error control. The proposed GP model, which takes into account influences of the source area, height difference between the head scarp crown and the base of the collapsed slope, average inclination angle of the source zone, and average slope angle of the travel path, was compared against six previous models from the literatures using 15 channelized rock avalanche events induced by the Wenchuan earthquake in the Tuojiang River Basin. The proposed GP model achieved significant improvement in estimating the channelized rock avalanche travel distance. In view of the limited number of channelized rock avalanche events, the application range of the proposed GP model is suggested. In conclusion, the proposed GP model could play a beneficial role in related rock avalanche disaster prevention.